CSAT (Customer Satisfaction Score)
A satisfaction metric measured by asking customers to rate a specific interaction or experience, typically on a 1–5 or 1–7 scale. CSAT is calculated as the percentage of customers who rate 4–5 (or 6–7), making it a transaction-level complement to NPS's relationship-level measurement.
CSAT (Customer Satisfaction Score) measures how satisfied customers were with a specific interaction — a support ticket, an onboarding session, a recent purchase. The standard question: "How satisfied were you with [X]?" with a 1–5 or 1–7 scale.
CSAT (%) = (count of 4–5 ratings / total responses) × 100
A CSAT of 85% means 85% of respondents picked the top two boxes. Industry benchmarks vary, but 80%+ is generally considered healthy across categories.
CSAT vs NPS
The two are complementary, not substitutes:
- CSAT measures this specific interaction — was that ticket resolved well, was that delivery smooth, was that onboarding clear?
- NPS measures the overall relationship — would you recommend us in general?
A customer can have high CSAT on a recent interaction but low NPS because they're dissatisfied with the product overall, and vice versa. Best-in-class support organizations watch CSAT per ticket; best-in-class CX organizations watch NPS quarterly.
Where CSAT is most useful
CSAT shines for transaction-level diagnosis — finding which support agents, channels, or interaction types produce dissatisfied customers. The granular feedback makes it actionable in ways NPS isn't. Many companies tie support team performance review to CSAT scores, which works as long as ratings aren't gameable (always include "what could have been better?" follow-up to discourage rating inflation).
Limitations
CSAT is recall-biased — customers rate their last interaction, not the average. It also suffers from non-response bias (people with mid-range opinions often don't respond). Pair CSAT trend lines with retention data to confirm satisfaction is translating to behavior.